1. Field of the Invention
The present invention relates generally to transaction fraud detection and more specifically to detecting credit card or other fraud by comparing voice samples to a database of known fraudulent voice signatures.
2. Introduction
Currently, many users complete transactions on the Internet with credit cards. Many other users hesitate to use credit cards on the Internet for fear of identity theft. Identity theft is when an imposter attempts to use another's personal information to complete a transaction, for example trying to purchase a DVD boxed set from Amazon.com using stolen credit card information. Because online transactions are not conducted face to face, many card-based methods, such as printing a photograph of the cardholder on the card itself, have failed to adequately protect from identity theft in the eyes of consumers and the media. Although the risks of identity theft are generally no greater on the Internet than in other settings, the public perception is that the Internet is a risky place for credit card transactions. This public perception slows commerce and prevents companies from generating billions of dollars of new revenue. A small percentage of repeat offenders can encourage the public perception of the Internet as a risky place for commerce. Accordingly, what is needed in the art is a way to improve fraud detection and increase public trust in electronic commerce.